MyoDat 8: EMG Data Acquisition & Analysis

Brief

Researchers, physiotherapists, and sports scientists needed a portable, high‑fidelity EMG recording system capable of operating in real‑world environments, on athletes, students, patients, and factory workers alike, while also providing a professional desktop analysis environment for transforming raw signal data into interpretable clinical and research outcomes.

Approach

MyoDat 8 was developed in C#/.NET with a WPF front end, providing the responsive user interface and real‑time rendering performance required for live, multi‑channel sEMG visualisation at high sample rates. Its architecture cleanly separates hardware communication, the real‑time acquisition pipeline, the analysis engine, and the visualisation layer, allowing new analysis routines to be introduced without changes to the data‑capture stack.

The live preview system can connect either to the telemetry receiver via USB or directly to the data loggers through an opto‑isolated USB interface, providing electrical isolation for user safety during connected operation. It displays scrolling waveforms for all active channels at full sample rate, enabling researchers to verify electrode placement and signal quality before recording begins.

A custom packet map and control protocol were also developed to support efficient and reliable transfer of signal data. The protocol incorporates error correction and is resilient to dropouts, including out‑of‑sequence frame errors, while maximising available bandwidth through support for per‑channel bit rates and variable sampling rates. In addition to signal data, the packet map periodically transmits device status information such as battery level and signal strength.

To support long‑duration recordings, a rapid‑loading multiresolution waveform pipeline was developed. Recordings are precomputed into peak‑and‑trough aggregates at multiple scales, ensuring that short‑duration events remain visible in overview displays while progressively finer detail is revealed as the user zooms in. These precomputed summaries can also accelerate downstream algorithms, such as threshold‑crossing detection. Together with indexed file access, lazy on‑demand loading, background I/O and processing, and viewport‑based rendering, this approach allows large multi‑channel recordings to be opened quickly and explored interactively without blocking the UI.

At the hardware level, the MT20 incorporates a hot‑swappable battery design, allowing continuous telemetric recording without interrupting acquisition. For offline recording, duration is limited only by onboard storage capacity. The DL8+ also supports an optional high‑capacity battery pack for extended recording sessions.

The software also supports video capture from a range of cameras via Microsoft DirectX, creating a visual record of the activity alongside the sensor data. The video stream is automatically synchronised using an IR/LED pulse emitted by the MT20 and DL8+ at the start of recording, aligning the video timeline with the physiological data. This removes the need for manual synchronisation and enables frame‑accurate overlay of signal traces on recorded footage.

The analysis engine processes signals through a configurable pipeline supporting raw waveform display, RMS envelope, rectified and integrated EMG, and frequency‑domain analysis, with user‑defined time‑base and amplitude scaling. Fatigue analysis compares spectral characteristics and median frequency across time windows within a session, while wavelet analysis extends the frequency‑domain toolkit beyond FFT for non‑stationary signals.

A gait analysis module combines EMG channels with electrogoniometer angle data and footswitch contact events. It automatically calculates cadence, stride length, step length, and joint range of motion, with left and right overlays providing an immediate visual comparison of gait patterns without the need for manual event marking.

An integrated Muscle Map encyclopaedia includes anatomical reference images and demonstration videos for each muscle group. Selecting a muscle displays the recommended electrode placement location alongside a video of the expected muscle action, helping to reduce setup errors and supporting teaching and training in academic and clinical settings.

A searchable subject database links subjects, sessions, and analysis results, while export to open formats ensures compatibility with third‑party statistical tools and publication‑ready workflows. Branded reports can incorporate organisational logos and be produced as printed documents, PDFs, or editable Word files.

My Role

• Technical Leadership: Led the software development team, overseeing feature delivery, bug fixes, release cycles, and key technology decisions across the product lifecycle for both the hardware driver stack and the MyoDat Windows application.
• Requirements, Specification & Architecture: Worked directly with physiotherapists, sports scientists, ergonomists, and clinical researchers to define requirements, write technical specifications for signal processing algorithms and hardware communication protocols, and shape the overall architecture spanning real-time acquisition, analysis, and visualisation.
• Signal Processing & Algorithm Development: Specified and implemented core EMG analysis algorithms including RMS envelope, rectified and integrated EMG, FFT spectral analysis, power spectral density with median frequency tracking, wavelet analysis, and the Spaepen quantified EMG method - enabling reliable fatigue assessment and muscle activity classification. Also led development of the gait analysis pipeline, combining EMG with electrogoniometry and footswitch data to automatically derive cadence, stride length, step length, and joint ranges of motion.
• Hardware & Device Integration: Led PC-side integration of the MT20 telemetry system and DL8+ data logger, including USB communication, MicroSD data transfer, live preview pipeline, video synchronisation via IR/LED pulse detection, and configuration of per-channel sampling rates, resolutions, and transducer types.
• Quality & Delivery: Directed testing and QA across signal fidelity validation, hardware communication reliability, and software acceptance testing. Managed versioning, installer packaging, and software distribution including the free 3-year upgrade programme.
• Stakeholder Engagement & Support: Collaborated with academic, clinical, and industrial partners to gather feedback, inform the product roadmap, produce user documentation, and deliver training and ongoing technical support.

Technical Specifications

Technology Stack

MT20 & DL8+ Hardware Devices

MT20 Transmitter: 16 analogue + 4 digital channels, 24-bit, 18 kHz DL8+ Data Logger: 8 channels + triaxial accelerometer, 72 × 55 × 18 mm

MyoDat 8 for Windows: Software

Live Preview: real-time signal quality check before recording Data Replay: scroll, slow-motion, compress or expand time-base Video Capture & Synchronisation: frame-accurate overlay via IR/LED pulse

Hardware Comparison: MT20 / DL20 / DL8+